Software valuations have reset, and the correction is holding. Across public and private markets the revenue multiples investors will pay for software have compressed: sharply in the public markets and steadily, over a longer stretch, in the private ones. For a software CEO, it cuts what the team’s equity is worth. For investors, it reshapes exit models and the returns any original entry price can produce.
PeakBridge reads the same reset as an opening. We focus on a specific corner of the market we call industrial software: the systems that run the operating layer of physical industries. In our case, this means the software embedded in and powering a healthy food system. Think of data-driven food waste reduction or quality control automation. More than 90% of the economy still sits in those real, tangible industries, and they spend roughly six times as much on services compared to software, and thirteen times as much on labor. Equipped with AI, software now has room to take on work that has long gone to consultants and headcount. That space is largest where adoption still lags, and food and beverage sits about a cycle behind technology and healthcare.
Few people have seen the problem from as many angles as Christopher Dughieri. He has been an investor, an advisor, an operator, and a customer, and has led roll-ups and acquisitions that built larger software companies. Most recently he was Chief Technology and Product Officer at SendCloud, the e-commerce delivery company founded in 2012 that has raised more than $100 million. Companies bring him in to help them work out AI, and he tends to leave their boards, investors, and leadership teams rethinking their core functions. At the PeakBridge Summit last month, Partner Thomas van den Boezem sat down with him to get his candid take on what it takes to build a winner in the category.
A quick lightning round to start. One word or one sentence – no nuance allowed. First: is software highly overvalued or undervalued at the moment?
It’s overvalued.
Next: are we going to see massive layoffs and crises or dramatic productivity improvement in traditional industries like food?
I think we’re going to see productivity gains.
And last: is the leadership of software companies well prepared to go through these challenges in the market or do they need to step up and work with their boards to fix glaring holes in their teams?
They have to step up. No question.
Ok, let’s go deeper. What is the market getting wrong about the value of software and about building a company in software in 2026?
Part of the problem is that money got cheap, and when it did, people stopped building clear ROI cases. That’s catching up with us now. Everywhere you hear ‘thanks to AI, we can cut the workforce by 30%.’ Most of that is a poor excuse for poor management that was there from the start – companies bloated without clear goals or the discipline to execute against them.
"Part of the problem is that money got cheap, and when it did, people stopped building clear ROI cases. That’s catching up with us now."
Christopher Dughieri
We see many software companies now sprinkling these artificial intelligence features on their existing solution, from simply replacing their website to something more structural. But when you go into these companies at a deeper level, what are some of the big red flags in terms of how they are actually set up technically?
I start higher up than most people expect. Strategy first, then vision – and then I want to look at the product roadmap together. The first red flag I normally get is roadmaps that are very optimistic and basically have no return on investment. There’s nothing there. And unfortunately I’ve seen it quite a lot. So I start there and normally I can stay there for a long time before I even touch the tech. Because tech is more of a how, and you have to start with the what and why. And I think a lot of times those are not clear enough.
To build on that, one of the specific stories we hear these days is ‘we have unique data.’ What’s the difference between a company that is indeed collecting all of this data and building a repository versus those that actually are able to turn it into intelligence and create moats with that data? What separates one from the other?
Static data is just data. It sits there. What matters is whether you’ve built a lifecycle around it; whether every customer interaction feeds something that gets smarter over time. That’s what compounds. That’s what creates something a competitor can’t simply replicate by signing up the same data provider.
“Static data is just data. It sits there. What matters is whether you’ve built a lifecycle around it; whether every customer interaction feeds something that gets smarter over time.”
Christopher Dughieri
We’re seeing companies increasingly engaging in mergers and acquisition as a way not just to expand revenues and market share but to bring in capabilities. You’ve been on the inside of some of those roll ups. What are some of the risk factors there? What is something the teams or boards get wrong?
The first question I always ask is: why are we doing this and what are we trying to achieve? What’s the deal rationale? I’ve seen cases where that wasn’t clear – and if it isn’t, you have no way to measure success after integration. That’s fundamental. Beyond that: I rarely buy tech to integrate tech. What I’m really after is the people – the ones who carry the domain knowledge and can keep developing the product over time. That principle has served us well across several transactions.
So with all of these macro dynamics in the market, what does it actually take for a board in 2026 to help their leadership teams build the right structures to make these decisions of equity hire versus integration?
It takes experience to ask the right questions, but it also takes boards willing to go further into the detail than they’re often comfortable with. Get crisp on the outcome. What is the return on this decision? I’ve sat with a portfolio company that had unlimited AI token spending. I’ve actually seen this.
“It takes experience to ask the right questions, but it also takes boards willing to go further into the detail than they’re often comfortable with. Get crisp on the outcome.”
Christopher Dughieri
And suddenly a CFO comes along and says, ‘the cost of the Anthropic license is increasing through the roof and we’re actually spending 50k per month.’ And then you ask, so what’s the return? What’s the value being created? And then on a board level, I think you have to be much more clear on certain principles and in how you invest early on. That’s just an example.
For investors reading this, if you were to perform due diligence on a software company, on an AI company to evaluate: do they really have the right talent? Is the technical foundation solid and scalable? Are they even set up from the higher level down to be able to integrate AI? How do you approach it and what are the red flags you look for that aren’t in a data room?
The first problem with the data room is that there’s rarely much data in it. You see a nice org chart. I go deeper right away: what’s the employee satisfaction score, and how has it trended over the last few years? I also search LinkedIn on key people myself – just to screen them. Then I look at GTM. What are the key performance metrics? Can you show me two years of actuals? ‘We don’t track that.’ That’s a problem. AI doesn’t fix a broken foundation. It just makes the cracks bigger, faster.
A lot of them talk about doing discovery on their own without asking the full GTM force. Why? There’s a lot of internal value to be captured from the people who actually meet customers every day. Also customer support and success. You have a lot of knowledge in these companies that is untapped. But sometimes founder led companies are very passionate about the product and it can be easy to go gut feeling on that. We need more product and not better product. And this is a shift that needs to happen a bit more in the software industry. It’s not always just more features.
“AI doesn’t fix a broken foundation. It just makes the cracks bigger, faster.”
Christopher Dughieri
I think there is no denying that whether right or wrong, there is going to be a bit of a shakeout. There is going to be consolidation where we’re going to see companies not make it. What can companies that are still at risk of being on the wrong side of this do to get back on the right side?
It comes back to fundamentals. Customers – actually understanding what they need and what they’ll pay for over time. Most companies lose track of that somewhere on the journey. The ones still standing will be the ones that kept their promises: to customers, to staff, to investors. The so-called SaaSpocalypse is largely a crisis of broken promises. Add financial discipline to that – scarcity creates the focus and creative pressure that abundance tends to kill. When you don’t have infinite money, you will make choices and you will find a way.